1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
|
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
//! The bulk of the HTML parser integration is in `script::parse::html`.
//! This module is mostly about its interaction with DOM memory management.
use document_loader::LoadType;
use dom::bindings::cell::DOMRefCell;
use dom::bindings::codegen::Bindings::ServoHTMLParserBinding;
use dom::bindings::global::GlobalRef;
use dom::bindings::js::{JS, Root};
use dom::bindings::refcounted::Trusted;
use dom::bindings::reflector::{Reflector, reflect_dom_object};
use dom::bindings::trace::JSTraceable;
use dom::document::Document;
use dom::node::Node;
use dom::text::Text;
use dom::window::Window;
use encoding::all::UTF_8;
use encoding::types::{DecoderTrap, Encoding};
use html5ever::tokenizer;
use html5ever::tree_builder;
use html5ever::tree_builder::{NodeOrText, TreeBuilder, TreeBuilderOpts};
use hyper::header::ContentType;
use hyper::mime::{Mime, SubLevel, TopLevel};
use js::jsapi::JSTracer;
use msg::constellation_msg::{PipelineId, SubpageId};
use net_traits::{AsyncResponseListener, Metadata};
use network_listener::PreInvoke;
use parse::Parser;
use script_task::{ScriptChan, ScriptTask};
use std::cell::Cell;
use std::default::Default;
use url::Url;
#[must_root]
#[derive(JSTraceable, HeapSizeOf)]
pub struct Sink {
pub base_url: Option<Url>,
pub document: JS<Document>,
}
impl Sink {
#[allow(unrooted_must_root)] // method is only run at parse time
pub fn get_or_create(&self, child: NodeOrText<JS<Node>>) -> Root<Node> {
match child {
NodeOrText::AppendNode(n) => Root::from_ref(&*n),
NodeOrText::AppendText(t) => {
let text = Text::new(t.into(), &self.document);
Root::upcast(text)
}
}
}
}
/// FragmentContext is used only to pass this group of related values
/// into functions.
#[derive(Copy, Clone)]
pub struct FragmentContext<'a> {
pub context_elem: &'a Node,
pub form_elem: Option<&'a Node>,
}
pub type Tokenizer = tokenizer::Tokenizer<TreeBuilder<JS<Node>, Sink>>;
/// The context required for asynchronously fetching a document and parsing it progressively.
pub struct ParserContext {
/// The parser that initiated the request.
parser: Option<Trusted<ServoHTMLParser>>,
/// Is this a synthesized document
is_synthesized_document: bool,
/// The pipeline associated with this document.
id: PipelineId,
/// The subpage associated with this document.
subpage: Option<SubpageId>,
/// The target event loop for the response notifications.
script_chan: Box<ScriptChan + Send>,
/// The URL for this document.
url: Url,
}
impl ParserContext {
pub fn new(id: PipelineId, subpage: Option<SubpageId>, script_chan: Box<ScriptChan + Send>,
url: Url) -> ParserContext {
ParserContext {
parser: None,
is_synthesized_document: false,
id: id,
subpage: subpage,
script_chan: script_chan,
url: url,
}
}
}
impl AsyncResponseListener for ParserContext {
fn headers_available(&mut self, metadata: Metadata) {
let content_type = metadata.content_type.clone();
let parser = ScriptTask::page_fetch_complete(self.id.clone(), self.subpage.clone(),
metadata);
let parser = match parser {
Some(parser) => parser,
None => return,
};
let parser = parser.r();
let win = parser.window();
self.parser = Some(Trusted::new(win.get_cx(), parser, self.script_chan.clone()));
match content_type {
Some(ContentType(Mime(TopLevel::Image, _, _))) => {
self.is_synthesized_document = true;
let page = format!("<html><body><img src='{}' /></body></html>",
self.url.serialize());
parser.pending_input.borrow_mut().push(page);
parser.parse_sync();
},
Some(ContentType(Mime(TopLevel::Text, SubLevel::Plain, _))) => {
// FIXME: When servo/html5ever#109 is fixed remove <plaintext> usage and
// replace with fix from that issue.
// text/plain documents require setting the tokenizer into PLAINTEXT mode.
// This is done by using a <plaintext> element as the html5ever tokenizer
// provides no other way to change to that state.
// Spec for text/plain handling is:
// https://html.spec.whatwg.org/multipage/#read-text
let page = format!("<pre>\u{000A}<plaintext>");
parser.pending_input.borrow_mut().push(page);
parser.parse_sync();
},
Some(ContentType(Mime(TopLevel::Text, SubLevel::Html, _))) => {}, // Handle text/html
Some(ContentType(Mime(toplevel, sublevel, _))) => {
if toplevel.as_str() == "application" && sublevel.as_str() == "xhtml+xml" {
// Handle xhtml (application/xhtml+xml).
return;
}
// Show warning page for unknown mime types.
let page = format!("<html><body><p>Unknown content type ({}/{}).</p></body></html>",
toplevel.as_str(), sublevel.as_str());
self.is_synthesized_document = true;
parser.pending_input.borrow_mut().push(page);
parser.parse_sync();
},
None => {
// No content-type header.
// Merge with #4212 when fixed.
}
}
}
fn data_available(&mut self, payload: Vec<u8>) {
if !self.is_synthesized_document {
// FIXME: use Vec<u8> (html5ever #34)
let data = UTF_8.decode(&payload, DecoderTrap::Replace).unwrap();
let parser = match self.parser.as_ref() {
Some(parser) => parser.root(),
None => return,
};
parser.parse_chunk(data);
}
}
fn response_complete(&mut self, status: Result<(), String>) {
let parser = match self.parser.as_ref() {
Some(parser) => parser.root(),
None => return,
};
parser.document.finish_load(LoadType::PageSource(self.url.clone()));
if let Err(err) = status {
debug!("Failed to load page URL {}, error: {}", self.url.serialize(), err);
// TODO(Savago): we should send a notification to callers #5463.
}
parser.last_chunk_received.set(true);
if !parser.is_suspended() {
parser.parse_sync();
}
}
}
impl PreInvoke for ParserContext {
}
#[dom_struct]
pub struct ServoHTMLParser {
reflector_: Reflector,
#[ignore_heap_size_of = "Defined in html5ever"]
tokenizer: DOMRefCell<Tokenizer>,
/// Input chunks received but not yet passed to the parser.
pending_input: DOMRefCell<Vec<String>>,
/// The document associated with this parser.
document: JS<Document>,
/// True if this parser should avoid passing any further data to the tokenizer.
suspended: Cell<bool>,
/// Whether to expect any further input from the associated network request.
last_chunk_received: Cell<bool>,
/// The pipeline associated with this parse, unavailable if this parse does not
/// correspond to a page load.
pipeline: Option<PipelineId>,
}
impl<'a> Parser for &'a ServoHTMLParser {
fn parse_chunk(self, input: String) {
self.document.set_current_parser(Some(self));
self.pending_input.borrow_mut().push(input);
if !self.is_suspended() {
self.parse_sync();
}
}
fn finish(self) {
assert!(!self.suspended.get());
assert!(self.pending_input.borrow().is_empty());
self.tokenizer().borrow_mut().end();
debug!("finished parsing");
self.document.set_current_parser(None);
if let Some(pipeline) = self.pipeline {
ScriptTask::parsing_complete(pipeline);
}
}
}
impl ServoHTMLParser {
#[allow(unrooted_must_root)]
pub fn new(base_url: Option<Url>, document: &Document, pipeline: Option<PipelineId>)
-> Root<ServoHTMLParser> {
let sink = Sink {
base_url: base_url,
document: JS::from_ref(document),
};
let tb = TreeBuilder::new(sink, TreeBuilderOpts {
ignore_missing_rules: true,
.. Default::default()
});
let tok = tokenizer::Tokenizer::new(tb, Default::default());
let parser = ServoHTMLParser {
reflector_: Reflector::new(),
tokenizer: DOMRefCell::new(tok),
pending_input: DOMRefCell::new(vec!()),
document: JS::from_ref(document),
suspended: Cell::new(false),
last_chunk_received: Cell::new(false),
pipeline: pipeline,
};
reflect_dom_object(box parser, GlobalRef::Window(document.window()),
ServoHTMLParserBinding::Wrap)
}
#[allow(unrooted_must_root)]
pub fn new_for_fragment(base_url: Option<Url>, document: &Document,
fragment_context: FragmentContext) -> Root<ServoHTMLParser> {
let sink = Sink {
base_url: base_url,
document: JS::from_ref(document),
};
let tb_opts = TreeBuilderOpts {
ignore_missing_rules: true,
.. Default::default()
};
let tb = TreeBuilder::new_for_fragment(sink,
JS::from_ref(fragment_context.context_elem),
fragment_context.form_elem.map(|n| JS::from_ref(n)),
tb_opts);
let tok_opts = tokenizer::TokenizerOpts {
initial_state: Some(tb.tokenizer_state_for_context_elem()),
.. Default::default()
};
let tok = tokenizer::Tokenizer::new(tb, tok_opts);
let parser = ServoHTMLParser {
reflector_: Reflector::new(),
tokenizer: DOMRefCell::new(tok),
pending_input: DOMRefCell::new(vec!()),
document: JS::from_ref(document),
suspended: Cell::new(false),
last_chunk_received: Cell::new(true),
pipeline: None,
};
reflect_dom_object(box parser, GlobalRef::Window(document.window()),
ServoHTMLParserBinding::Wrap)
}
#[inline]
pub fn tokenizer(&self) -> &DOMRefCell<Tokenizer> {
&self.tokenizer
}
}
impl ServoHTMLParser {
fn parse_sync(&self) {
// This parser will continue to parse while there is either pending input or
// the parser remains unsuspended.
loop {
self.document.reflow_if_reflow_timer_expired();
let mut pending_input = self.pending_input.borrow_mut();
if !pending_input.is_empty() {
let chunk = pending_input.remove(0);
self.tokenizer.borrow_mut().feed(chunk.into());
} else {
self.tokenizer.borrow_mut().run();
}
// Document parsing is blocked on an external resource.
if self.suspended.get() {
return;
}
if pending_input.is_empty() {
break;
}
}
if self.last_chunk_received.get() {
self.finish();
}
}
fn window(&self) -> &Window {
self.document.window()
}
}
impl ServoHTMLParser {
pub fn suspend(&self) {
assert!(!self.suspended.get());
self.suspended.set(true);
}
pub fn resume(&self) {
assert!(self.suspended.get());
self.suspended.set(false);
self.parse_sync();
}
pub fn is_suspended(&self) -> bool {
self.suspended.get()
}
}
struct Tracer {
trc: *mut JSTracer,
}
impl tree_builder::Tracer for Tracer {
type Handle = JS<Node>;
#[allow(unrooted_must_root)]
fn trace_handle(&self, node: JS<Node>) {
node.trace(self.trc);
}
}
impl JSTraceable for Tokenizer {
fn trace(&self, trc: *mut JSTracer) {
let tracer = Tracer {
trc: trc,
};
let tracer = &tracer as &tree_builder::Tracer<Handle=JS<Node>>;
let tree_builder = self.sink();
tree_builder.trace_handles(tracer);
tree_builder.sink().trace(trc);
}
}
|
